Legacy of Early Anthropogenic Effects on Recent Lake Eutrophication

Home , French Prealps, Preboreal

Anthropocene 24 (2018) 72–87

Contents lists available at ScienceDirect

Anthropocene

journa l homepage: www.elsevier.com/locate/ancene

Legacy of early anthropogenic effects on recent lake eutrophication

(Lake Bénit, northern French Alps)

a,b, c c c

Manon Bajard *, David Etienne , Sébastien Quinsac , Etienne Dambrine ,

b c c b

Pierre Sabatier , Victor Frossard , Jérémie Gaillard , Anne-Lise Develle ,

b b c

Jérôme Poulenard , Fabien Arnaud , Jean-Marcel Dorioz

E.A. 6293 GeHCO, University of Tours, 37000 Tours, France

Univ. Grenoble Alpes, Univ. Savoie Mont Blanc, CNRS, EDYTEM, 73000 Chambéry, France

UMR INRA 42 CARRTEL, Université Savoie Mont Blanc, 73376 Le Bourget-du-Lac, France

A R T I C L E I N F O A B S T R A C T

Article history: Mountain lakes are integrated sentinels of changes in the terrestrial environment, where these changes

Received 4 June 2018

threaten the quality of the ecosystem services these lakes provide, including high biodiversity, economic

Received in revised form 22 November 2018

and leisure activities. Few evidentiary records exist of the long-term relationships between human

Accepted 23 November 2018

pressure and observed impacts. Multiproxy analyses of the Lake Bénit sediment sequence, including

Available online 28 November 2018

dating, grain-size, geochemistry, pollen, non-pollen palynomorphs and chironomid assemblage

reconstructions, allowed reconstruction of past environmental evolution and lake trophic changes.

Keywords:

Combined with soil analyses of the catchment, these data provide a record of the relationships between

Phosphorus

Grazing human activities and the lake-catchment ecosystem, and show the effect of inundation of the shore

previously used as pasture. From 2100 to 110 0 yrs cal. BP, the catchment was forested. During the Middle

Water-level ﬂuctuation

Lake sediment Ages, grazing deforested the catchment, triggering an increase in erosion and a change in sediment

Soil sources. The lake remained oligotrophic over most of the last millennia. The trophic state changed

Erosion abruptly in the 20th century with intensiﬁcation and multiplication of tourist activities in the catchment,

i.e., ﬁshing, hiking, while pastoral activities decreased. The sudden eutrophication coincides with an

artiﬁcial increase of the lake water level in AD 1964 to improve ﬁshing activities. A release of phosphorus

(P) from the ﬂooded soils was observed, which may be responsible for the current eutrophication. One

thousand years of grazing practices would have led to the observed P concentrations in the soils of the

lake shore, transferred by the cattle to this area. Our study highlights the combined effects of past and

recent activities on the current eutrophication process, and the legacy of both soils and early

anthropogenic activities.

1. Introduction the quality of water and recreation areas). The intense use of these

services may disrupt their functioning and harm life.

Understanding the effects of past activities on both the current Lacustrine ecosystems are very sensitive to environmental

status/structure of an ecosystem and its functioning is challenging changes, especially in mountains and boreal areas where human

but necessary to ensure the sustainability of the “Earth critical activities and climate change appear to be key factors in the

zone” and restore damaged ecosystems (Reed-Andersen et al., disruption of these ecosystems (Rogora et al., 2003; Brisset et al.,

2000; Costanza et al., 2007; Arnaud et al., 2016; Dubois et al., 2013; Bajard et al., 2017a). Lakes are considered to be integrated

2018). Ecosystems of the critical zone (i.e., where human societies sentinels of local and global environmental changes, as they

develop at the interface between the atmosphere, hydrosphere, accumulate both the direct effects of environmental forcing factors

biosphere and geosphere), such as mountain lake environments, on the lake itself and those on its watershed (Carpenter et al., 2007;

provide services for life (e.g., food supply, climate regulation and Williamson et al., 2008; Adrian et al., 2009; Schindler, 2009). The

morphometric attributes (e.g., shape, depth) of lakes, which

determine their oxygenation, especially in benthic areas, are very

sensitive to energy inputs (thermal regimes and runoff), and the

* Corresponding author at: E.A. 6293 GeHCO, University of Tours, 37000 Tours,

France. inputs from the catchment are regulated by the same hydrody-

E-mail address: [email protected] (M. Bajard). namic processes. They control trophic states of lakes, especially

https://doi.org/10.1016/j.ancene.2018.11.005

M. Bajard et al. / Anthropocene 24 (2018) 72–87 73

through nutrients inputs. In this context, the littoral zone appears human activities and their consequences for soil erosion related to

to be a key factor in the relationship between the effects on the forest clearing and cattle or sheep grazing since the early Neolithic

catchment and those on the trophic web. The sensitivity of lakes to period in the French Alps (e.g., Giguet-Covex et al., 2011, 2014,

disruption is also the consequence of the hydrological character- Arnaud et al., 2012, 2016; Brisset et al., 2012; Simonneau et al.,

istics of aquatic ecosystems, such as rapid overturn, mixing and 2013). Lake sediments allow us to reconstruct changes in the lake

resistance, as well as the adaptation of aquatic communities to trophic status. The analysis of the geochemical P contents, of

modiﬁcations of their environment. diatoms or chironomid assemblages from lake sediments is used as

Eutrophication is one of the most important threat to tracers of nutrient inputs and oxygenation.

freshwater ecosystems due to the degradation of water quality, We focused on Lake Bénit (1450 m a.s.l.), which is located in the

and thus the quality of the entire ecosystem, due to the potential subalpine belt in the western French pre-Alps and has been

loss of habitats and biodiversity and other associated ecosystem identiﬁed by a local limnological survey as a “lake to protect”

services (Dudgeon et al., 2006; Davidson and Jeppesen, 2013). (Druart et al., 1999) due to its eutrophication processes. The lake

Eutrophication can be a natural process describing an increase in catchment includes both rocky slopes and pastures that are

primary production as a result of an increase in nutrient inputs. currently partly recolonized by trees (mainly spruce) and shrubs

This process occurs naturally with the gradual ﬁlling of lakes, (Rosa sp.) The spot is highly frequented by hikers and anglers in

ponds and other water reservoirs, or in the case of drying and lake summer. Therefore, a summer bar was established in CE 1952, and

level decrease, in response to climate change. Eutrophication can the water lake level was intentionally increased in CE 1964. These

also result from excessive anthropic inputs of nutrients such as recent changes in anthropic pressures applied on the lake-

nitrogen (N) and phosphorus (P) which increase the primary catchment system have made this area an ideal location for our

production (Carpenter et al., 1995; Correll, 1998; Bennett et al., study to (i) reconstruct the past landscape evolution associated

2001; Jenny et al., 2013; Pinay et al., 2017) and so-called “human with human activities over the last millennia, (ii) identify the

induced-eutrophication” or “cultural eutrophication” (Schindler, sources of sediments and erosion dynamics and (iii) understand

2012). In lakes, eutrophication is almost always induced by an the links between past and present human activities and the

excess of phosphorous that is generally provided by an external trophic state of the lake.

load from the watershed (Schindler et al., 1995; Schindler, 2012)

and often increases suddenly (i.e., within a few years). 2. Material and methods

External loads causing eutrophication often result from human

activities that modify the natural biogeochemical cycling of 2.1. Study site

nutrients within lake watersheds and tend to increase the transfer

from the watershed to the lake (e.g., Vitousek, 1997; Smith et al., Lake Bénit (1450 m a.s.l.) is located in the western Alps in the

1999). Agriculture, and even less extensive pastoral activities, are Bargy massif and is part of Marnaz and Mont-Saxonnex villages. It

mainly associated with the redistribution of nutrients in soils. is located between Annecy and Geneva on the left side of the Arve

Grazing can produce strong heterogeneities in soil nutrients, Valley (Fig. 1-A). The lake, which is of glacial origin, is delimited

concentrating N and P through droppings in small areas such as ﬂat by a limestone cliff of Cretaceous age in the south that rises up to

resting places, near water, or in milking places within a few 2230 m a.s.l. and by a Würm moraine ridge in the west that rises

centuries (Bennett et al., 2001; Jewell et al., 2007; Williams and up to 1570 m a.s.l. (Fig. 1-B). The surface area of the lake is 0.04

Haynes, 1990). This change in the spatial pattern of soil nutrients km , and its maximum depth reaches 8.7 m (Sesiano, 1993). The

can contribute to an increase in the ﬂuxes transferred to surface lake is fed by four springs ﬂowing through the moraine and the

waters if these enriched source areas are connected to surface base of the scree. It drains a catchment area of 0.9 km (Druart

waters. Pastoralism has been present for millennia in the Alps, et al., 1999). A ﬂysch area composed of schists, sandstones and

from lowlands to subalpine areas, exerting long-term effects on limestones of Nummulitic age crops out between the Cretaceous

ecosystems, such as changes in vegetation and soil (Giguet-Covex limestone cliff and the moraine ridge as gullies (Sesiano and

et al., 2014; Pini et al., 2017; Bajard et al., 2017b). Soils have a Muller, 1986). The ﬂysch area is a ﬂat herbaceous zone that is

memory, i.e., they recorded information of their evolution and accentuated and gullied upstream (Fig. 1-B–D). The lake is

land-use evolution, and can retain the impacts of past and present covered in ice every winter and protected from the sun by the

human activities which can be chronologically recorded in lake cliff. The mean annual temperature of the village of Mont-

sediment archives through erosion (Mourier et al., 2010; Sabatier Saxonnex is 7.6 C and the annual precipitations is 1110 mm

et al., 2014; Poulenard et al., 2015; Bajard et al., 2016). Erosion and, (climate-data.org). Snow covers the area during winter, and

more generally, all modiﬁcations of external inputs to freshwater avalanches regularly go down the active scree. The east side of the

reservoirs, can result in changes in limnology. However, the catchment is forested, while the moraine ridge is partially

interactions between past and recent environmental modiﬁcations forested and pastured (Fig. 1-C,D). A small bar was built next to

(whether or not of anthropic origin) are more difﬁcult to assess but the lake in 1952 and the lake level was increased by 2 m in 1964 to

necessary to understand the functioning of current environments. facilitate ﬁshing activities (Fig. 2).

It is necessary to better understand the relationships between The temperatures of the lake range from 15 to 18 C at the

different pressures on lakeshores and ecosystem responses (Coops surface to 8–10 C at the bottom in August (measured in 2013 and

et al., 2003; Ostendorp et al., 2004; Dubois et al., 2018). In this 2016) and from 0 C at the surface to 4 C at the bottom in spring

context, we focus on the human impact on both lakeshore soils and (measured in March 2014 and April 2017). Lake mixing occurs

lake trophic status; we especially focus on the link between human when the water column overturns in the autumn when the

À1

pressures that can last for several millennia (e.g., agriculture) and surface cools, thus oxygenating the bottom (8 mgO2.L ). The lake

th À1

have intensiﬁed over the 20 century (Vitousek, 1997; Steffen exhibits anoxia during most of the year (0 mgO2.L in spring and

et al., 2015) and the development of eutrophication. To consider summer at the bottom), and the oxycline can reach depths of 2 to

previous modiﬁcations of the environment, our approach required 5 m (data from August 2013 and March 2014). The PO4

À1

to study lake-catchment systems together on long time scales from concentrations range from 5 to 80 mg.L , and a Secchi disk

a paleoenvironmental perspective. disappeared at a water depth of approximately 3 m, indicating

Paleoenvironmental studies based on lake sediment archives that the bottom water exists in a mesotrophic to eutrophic state

allow us to reconstruct the evolutions of past landscapes and (Nürnberg, 1996; Wetzel, 2001).

74 M. Bajard et al. / Anthropocene 24 (2018) 72–87

Fig. 1. Location of Lake Benit in the Western Alps a), simpliﬁed geological map of the catchment b), aerial photograph c) and view from the west side of the lake d). The

bathymetry of the lake is adapted from Sesiano (1993).

2.2. Soil sampling depth of 6 m. This core is considered to be the reference master

core. A second coring campaign from the ice-covered surface in

Four soil proﬁles (BENS1, BENS3, BENS4 and BENS5) were winter 2016 allowed us to retrieve one core, that was 165 cm long,

described and sampled based on their horizons (Figs. 1-B and 2). at a depth of 8 m (BEN16-P1 - IGSN in progress). All cores were split

The latest ﬁne ﬂow of the limestone scree was also sampled lengthwise into two halves. Each half-section was described in

À1

(BENS2), as were the coarse elements of the ﬂysch (BENS4-R). The detail, and pictures were taken with a resolution of 20 pixel. mm .

FAO (Food and Agriculture Organization) soil classiﬁcation (WRB - The lithological description of the cores allowed the identiﬁcation

FAO, 2014) and the Guidelines for soil description (FAO, 2006) were of different sedimentary units. Correlations between cores BEN14-

used for horizon and soil denominations. Four transects, located P4, BEN14-P3 and BEN16-P1 were performed using visual and XRF

perpendicular to the lake moraine shore (i.e., glacial deposit in core scanner analyses (See SI-I).

Fig. 1), were auger-sampled every 15 cm to 45 cm when possible,

with three positions: out of the lake (A), in the current littoral zone 2.3.2. Grain-size distribution

(B) and in the lake (C, < 50 cm of water), corresponding to soils The grain-size distribution of the sediment was determined

ﬂooded by changes in water level (Fig. 3). The samples collected every centimeter using a Malvern Mastersizer 2000 G laser particle

from soil proﬁles and transects were dried and sieved to a size of sizer after its organic matter (OM) was burned by hydrogen

2 mm for further laboratory analyses. The coarse elements from the peroxide during a three-day period. Ultrasounds were used to

ﬂysch and the scree were crushed for geochemical analysis. dissociate mineral particles and to avoid their ﬂocculation. Every

10 cm, several drops of a 0.5 N-HCl solution were added to assess

2.3. Sediment sequence and chronology the effect of decarbonatation on the grain-size. The results of the

grain-size distribution were processed using MATLAB software and

2.3.1. Coring presented in a contour plot with a color scale based on the

The ﬁrst coring campaign performed in September 2014 abundance of particles (in percentage) in each grain-size class

allowed extracting four short gravity cores from the lake at depths (Fig. 4).

of eight meters depth (BEN14-P1 – IGSN no: IEFRA05HO; BEN14-P2

- IGSN no: IEFRA05HP; BEN14-P3 - IGSN no: IEFRA05HQ) and 6 m 2.3.3. Analysis of pollen and non-pollen palynomorphs

(BEN14-P4 - IGSN no: IEFRA05HR - IGSN codes refer to an open A total of 26 subsamples of approximately 1 cm were collected

international database, www.geosamples.org). The cores from the from the BEN14 P4 core for the analysis of pollen and Non-Pollen

deepest part of the lake were 65 to 78 cm long, while the core at a Palynomorphs (NPPs). After adding tablets of exotic spores

depth of 6 m was 98 cm long. Thus, most of the analyses were (Lycopodium clavatum) (Stockmarr, 1971) to calculate concentra-

performed on the longest core BEN14-P4, which was collected at a tions of NPPs, subsamples were prepared chemically (HCl, NaOH,

M. Bajard et al. / Anthropocene 24 (2018) 72–87 75

Fig. 2. Pedologic description of the soil proﬁles (texture, structure, color), including OM content (Loss On Ignition at 550 C) and measured pH values. Horizons were deﬁned

using the WRB (FAO, 2014) and the Guidelines for soil description (FAO, 2006).

76 M. Bajard et al. / Anthropocene 24 (2018) 72–87

Fig. 3. a) Boxplot of available phosphorus (POlsen) distribution along the four transects on the Lake Bénit shore, out of the lake (A), in the current littoral zone (B), in the lake

(C, < 50 cm of water) and according to depth. b) Distribution of POlsen concentrations around the lake in the surface samples (0–15 cm). c) Distribution of total phosphorus

(PTot) concentrations around the lake in the surface samples (0–15 cm) and in the surface samples of BENS3, BENS4 and BENS5 soils.

HF, HCl and acetolysis) and physically (any material greater to analysis to identify ecological changes within the lake. Speciﬁcally,

100 mm was removed by sieving) following the standard procedure chironomid reconstruction allowed us to infer hypolimnetic

of Faegri and Iversen (1989). At least 500 pollen grains of terrestrial oxygen conditions (Brodersen and Quinlan, 2006) as well as

plants (Total Land Pollen, TLP) were identiﬁed and counted in each habitat availability (e.g., Langdon et al., 2010). The samples were

sub-sample. Pollen identiﬁcation was based on an identiﬁcation exposed for successive 2 h periods in HCl (10%) and KOH (10%)

key (Beug, 2004) and photograph book (Reille, 1992). Grass pollen solutions at room temperature to dissolve carbonates and

grains greater than 40 lm in annulus diameter were classiﬁed as deﬂocculate OM, respectively. The residues were sieved through

Cerealia-type to exclude wild grass species (Beug, 2004). Arboreal a non-standard 150-mm-mesh-size ﬁlter. Chironomid head cap-

(AP) and Non-Arboreal Pollen (NAP) were expressed as percen- sules (HC) were handpicked from the sieving residue under a

tages of the total land pollen sum (TLP) to reconstruct the stereomicroscope at a 40 Â magniﬁcation. The HC were then

vegetation dynamic in the catchment. mounted on slides using Aquatex (Merck, Darmstadt, Germany)

NPPs were counted in pollen slides following the procedure mounting medium, and identiﬁcation was performed under a

described by Etienne and Jouffroy-Bapicot (2014) and expressed in compound microscope at 200–1000 Â magniﬁcation. Identiﬁca-

À3

terms of concentration (spores. cm ) and accumulation rates tion was based on the guides of Brooks et al. (2007). Due to the use

-2 -1

(spores. cm . year ). Among all of the NPPs identiﬁed, the strict of a non-standard mesh size (150 mm instead of more commonly

coprophilous fungal ascospores Sporormiella sp. (HdV-113) and used - 100 mm), some bias might be present in our analysis and

Podospora sp. (HdV-368) (Van Geel, 2002) were summed and used results were therefore interpreted accordingly.

as an indicator of local variations in grazing pressure (Etienne et al.,

2013; Doyen and Etienne, 2017). 2.3.5. Chronology

The chronology of the Lake Bénit sediment sequence is based on

2.3.4. Chironomid analysis seven C measurements performed on terrestrial plant macro-

210

A total of 74 one-centimeter-thick samples were continuously remains, as well as short-lived radionuclide measurements ( Pb

137

sampled along the BEN14-P3 sediment core for chironomid and Cs). AMS Radiocarbon dates were performed using an

M. Bajard et al. / Anthropocene 24 (2018) 72–87 77

Fig. 4. Comparison of the downcore sediment grain-size data of Lake Bénit with loss on ignition at 550 C and 950 C, its NCIR (Non-Carbonate Ignition Residue) and the XRF

geochemistry of major elements (CaO, SiO2, K2O, TiO2 and K2O/TiO2). The color scale of the grain-size contour plot refers to the abundance (in percentage) of each grain-size

class, with the less abundant in blue and the most abundant in red (For interpretation of the references to colour in this ﬁgure legend, the reader is referred to the web version

of this article).

accelerator mass spectrometer (AMS) at the Poznan Radiocarbon 2 cm) were performed through a 4-mm-thick Ultralene ﬁlm in a

Laboratory and at the Beta Analytic Radiocarbon Dating Laboratory 32-mm-diameter plastic cup. The samples were triplicated and

in London (Table 1). The C ages were converted to ‘calendar’ years analyzed over 60 s by using the internal calibration mode of the

using the calibration curve IntCal13 (Reimer et al., 2013). Short- GeoChem Standard instrument (Shand and Wendler, 2014). The

lived radionuclide measurements were performed at the Modane OM content that was deduced from the LOI at 550 C was added to

Underground Laboratory (LSM) on the uppermost sediments the sum of the major elements. Consequently, the elemental

(Reyss et al., 1995) using a 2-cm sampling step. The age–depth results were brought to 100% to overcome the closed sum effects

model was then generated using R software and the R-code that are linked to variations in OM (Baize, 2000). The standard

package ‘Clam’ version 2.2 (Blaauw, 2010). deviations of the replicates were lower than the errors introduced

by the instrument, which were thus used as conservative

2.4. Soil and sediment analysis measurement uncertainties.

2.4.1. Loss on ignition 2.4.3. CNP analysis

Loss on ignition (LOI) analyses were performed on all samples The total organic carbon (Corg) and nitrogen contents of lake

from soil transects and proﬁles, according to depth, to estimate the sediments were determined through combustion with an elemen-

organic matter (OM) and carbonate contents following heating at tal CHN analyzer every 2 cm until reaching a depth of 22 cm and

550 C for 4 h and 950 C for 2 h, respectively (Heiri et al., 2001). LOI then every 3 cm below, on samples previously decarbonated with

analyses were also performed on sediment samples obtained using an acid solution. Then, the C/N ratio was calculated to determine

a continuous 1-cm sampling step along the entire sediment the origin of OM. The total phosphorus concentration (Ptot) was

sequence, with the same heating parameters. measured in the surface horizons of the BENS3, BENS4 and BENS5

soils, in the 0-15-cm samples of shore-lake transects and in the

2.4.2. P-XRF mineral geochemistry sediment sequence after undergoing hot sodium hydroxide

A portative ED-XRF (P-XRF) spectrometer (S1 TITAN Bruker) digestion. A labile fraction of total P (Polsen) was extracted from

was used to measure the major element contents, which were soil transects, every 15 cm according to depth, and sediments

expressed as relative percentages of oxides. The analyses of gently according to the method of Olsen (1954) by mixing 1 g of sediment

crushed soil samples (according to horizons), soil transect samples (or soil) with 20 ml of a 0.5 N-NaHCO3 solution, following the same

(every 15 cm according to depth), and sediment samples (every sampling for sediment samples as used for Corg and N analyses. This

Table 1

Radiocarbon ages for the Lake Bénit sediment sequence.

Core name Lab. Code Depth (cm) Sample type Age (yrs BP) Cal. Min (cal. BP) Cal. Max (cal., BP) Prob.

BEN14-P4 Poz-69616 42,0 Plant macroremains 730 Æ 30 653 709 91,5

BEN14-P4 Beta-46357 48,0 Plant macroremains 1270 Æ 30 1173 1286 92.3

BEN14-P4 Beta-462835 59,5 Wood, herbs 1120 Æ 30 956 1086 91,8

BEN16-P1 Poz-81886 63,5 Seed 440 Æ 30 462 532 94,1

BEN14-P4 Poz-69617 71,5 Bark 1150 Æ 30 980 1150 86,8

BEN16-P1 Poz-81887 96,7 Twigs, wood 2060 Æ 30 1948 2116 95

BEN14-P4 Poz-69615 98,0 Wood 2335 Æ 30 2312 2436 95

Bold samples correspond to dates excluded from age-depth modeling. *

XRF-correlated depth.

78 M. Bajard et al. / Anthropocene 24 (2018) 72–87

fraction can be considered to be representative of the bioavailable c). The Ptot concentrations are higher in the soils that developed on

À1

fraction of P, both in soils and sediments (Jordan-Meille et al., the ﬂysch area (> 700 mg.kg ) in the southwest region of the lake

À1

1998). The spectrophotometric determination of P-PO4 in all than those on the moraine ridge soils (< 700 mg.kg ), and they are

À1

extracts was carried out using the molybdenum blue method much lower in the moraine ﬂooded soils (< 500 mg.kg ).

(Murphy and Riley, 1962) at 882 nm on a Cary 50 UV–vis

spectrometer (Agilent). 3.3. Lithology and sediment analyses

Photosynthetic pigments were extracted overnight using a

solution of acetone and water (90:10) (76 samples collected at 1 to Based on lithological descriptions and grain-size and LOI data,

2-cm sampling steps from the BEN14 P4 core). The resulting three sediment units were identiﬁed in BEN14-P4 (Fig. 4). Unit 3,

extracts were used to quantify the total carotenoids via a which occurs between 98 and 65 cm, is very dark gray and presents

spectrophotometer following the recommendations of Guilizzoni an important sedimentary fraction between 20 and 100 mm, with

et al. (2011). The carotenoid concentrations were expressed in mg limestone gravels that have been identiﬁed as dropstones. The

À1 À1

per gram of sediment per year (TC in mg. g .an ) as an indicator sediment of Unit 2 (65 to 13 cm) is greenish-gray, with millimetric

of the variation in the lacustrine primary production and then laminae. Unit 1 (top 13 cm) is very dark gray with a darker level

À1

normalized by the organic matter content (mg. g of OM) to appearing between 13 and 11 cm and very homogenous sediment

reconstruct the past phosphorus concentration in the water in the ﬁrst six centimeters. The grain sizes in Units 2 and 1 mainly

À1

column (TPw in mg.L ) following the formula published by range between 2 and 10 mm but are more dispersed in Unit 1

Guilizzoni et al. (2011). (Fig. 4). The LOI at 550 C range from 15 to 35%, 8 to 15%, and 14 to

19% in Units 3, 2 and 1, respectively (Fig. 4). The LOI at 950 C range

2.5. Statistical analysis from 2 to 21% in Unit 3 and remains less than 6% in Units 1 and 2.

The CaO content shows the same trend as the LOI 950 C, with

Principal component analysis (PCA) was performed with the higher variability in Unit 3 and very low contents in Units 2 and 1.

results of the geochemical analyses of the lake sediments and soil, The contents of SiO2, K2O and TiO2, as well as K2O/TiO2 ratio

soil transect and rock samples of the catchment, including the TiO2, present the same variations as the NCIR of LOI, with lower contents

SiO2, K2O, Al2O3, CaO, Fe2O3 contents, K2O/TiO2, Al2O3/TiO2 ratios and higher variability in Unit 3 and higher but less variable

and OM content. PCA was used to determine the origin of the contents in Units 2 and 1 (Fig. 4).

sediment in the catchment (Sabatier et al., 2010). These analyses

were performed using R 2.13.1 (R Development Core Team, 2011). 3.4. Age-depth model

3. Results The measurement of short-lived radionuclides allowed the

dating of the upper 19 cm of the core (Appleby and Oldﬁeld, 1978).

210

3.1. Soil characterization A logarithmic plot of Pbex activities (Fig. 5-a) according to mass-

depth shows stable values for the upper 8 cm followed by a linear

Two main soil types were identiﬁed in the Lake Bénit catchment decrease (Fig. 5-a). According to the ‘constant ﬂux, constant

based on their main evolutionary processes (e.g., humiﬁcation, sedimentation rate’ (CFCS) model (Goldberg, 1963; Krishnaswamy

210

decarbonatation and acidiﬁcation), pH values and parent material. et al., 1971), the Pbex activities indicate a mean accumulation

À3 À1

Folic Umbrisols develop on stabilized limestone screes on the right rate of 2 mg.cm .yr between 19 and 6 cm and very fast sediment

210

shore of the lake (BENS1), while Cambisols develop on ﬂysch deposition for the ﬁrst 6 cm with almost constant Pbex activities.

À3 À1

(BENS3 and BENS4) and on the moraine ridge (BENS5). BENS1 is If we use this accumulation rate (70 mg.cm .yr ), the ﬁrst 6 cm

137

characterized by a shallow depth (< 25 cm), a very high OM content represent 2 years. The Cs activity proﬁle (Fig. 5-a) shows a ﬁrst

(> 50%) and pH values ranging from 5 at the surface to 7 at the peak at a depth of 13 Æ1 cm and another one at 9 Æ 1 cm; these

bottom (Fig. 2). Cambisols are deeper and mainly comprise silty peaks are interpreted to reﬂect the maximum atmospheric

137

clay and OM contents of less than 20% (Fig. 2). The pH of BENS3 is 8 production of Cs in 1963 (nuclear tests) and the Chernobyl

and ranges with depth from 6.5 to 7.5 for BENS4 and from 6 to 7 for accident in 1986 (Appleby et al., 1991), respectively. The beginning

137

BENS5. The texture of BENS3 is coarser than those of the other of the ﬁrst Cs activity peak could correspond to the ﬁrst fallout of

137

Cambisols because it developed in the alluvial zone before the lake, Cs related to nuclear weapons tests dated to CE 1955 in the

137

downstream of the ﬂysch gullies. BENS3 and BENS4 present some Northern Hemisphere. Considering the Cs peaks in 1986 and

rust patterns (Fig. 2). The total phosphorus contents in the surface 1963, the sedimentation rate determined from the extrapolation of

À1 210

horizons of BENS, BENS4 and BENS5 are 895, 815 mg.kg and Pbex activity to the top of the core does not agree with the coring

À1 137 210

647 mg.kg , respectively. year (2014). Both visual observations and the Cs and Pbex

activities of the ﬁrst six centimeters of sediment indicate the

3.2. Distribution of phosphorus along the soil transects strong homogeneity of the sediment that could be an artefact out

of the continuous sedimentation. This hypothesis is supported by

In the soils around the lake, the POlsen concentrations range the comparison with the other cores, where these 6 cm are not

À1 À1

from 1.7 to 10 mg.kg (5.1 mg.kg on average) in the ﬁrst 15 cm recorded (Fig. 5-b). A problem during the coring could have

À1

and remain lower than 2 mg.kg below this depth (Fig. 3-aA). In occurred, if the tube had touched the bottom of the lake and

the littoral zone, POlsen ranges from approximately 0.8 to 7.5 mg. bounced before the real coring started. Slumps are also common in

À1 À1 À1

kg (3.4 mg.kg on average) in the surface, from 2 to 3 mg.kg lake sedimentary deposits, especially on the shores of lakes.

À1

between 15 and 30 cm and less than 3 mg.kg below this depth Erosion of the shores could also have bring punctual clods of soil

À1

(Fig. 3-aB). The POlsen concentrations range from 2 to 6.8 mg.kg and sediments. Thus, the ﬁrst 6 cm of BEN14-P4 were not

À1

(4.5 mg.kg on average) in the surface and remain very low until considered in the subsequent analyses. Following this correction,

30 cm in the ﬂooded soils (Fig. 3-aC). The distribution of the the artiﬁcial radionuclide peaks and ages derived from the

bioavailable P concentration in the 0–15 cm layer decreases from 210Pbex-CFCS model show good agreement and provide an

the shore (A) to the lake (C), except in transect 4 (Fig. 3-b). accurate and continuous age–depth relationship over this part

The concentrations of total phosphorus (Ptot) on the surface (0– of the sediment sequence. These data were then incorporated into

À1

15 cm) of the soil transects range from 360 to 808 mg.kg (Fig. 3- Clam to generate an age–depth model of the entire core (Fig. 5-b).

M. Bajard et al. / Anthropocene 24 (2018) 72–87 79

Fig. 5. Age-depth model of the Lake Benit sedimentary sequence including 210Pb/137Cs chronology (a) and 14C ages (b). Uncertainties are included in the 137Cs activity dots.

The top core of BEN14-P4 was corrected for its ﬁrst 6 cm to ﬁt with the 210Pb/137Cs chronology.

À1

A total of 7 radiocarbon ages were obtained from the selected cal. BP, with a mean sedimentation rate of 0.25 mm.yr until

macroremains collected from the BEN14-P4 and BEN16-P1 cores 1200 cal. BP. Unit 2 covers the period between 1000 and 100 yrs cal.

(Table 1). Three radiocarbon dates were excluded from the age– BP. Unit 2 exhibits a change in the sedimentation rate at its base,

À1

depth model (bold in Table 1). The Poz-69615 and Poz-81887 where it increases to 1.5 mm.yr . Then, it decreases from 1.5 to

À1

samples were too close to each other in terms of both depth and 0.35 mm.yr and ﬁnally increases again after 400 yrs cal. BP,

age. Thus, the older sample (Poz-69615) was removed from the mainly over Unit 1, until reaching its present day value of 1.6 mm.

À1

age-depth modeling. The Poz-81886 sample, which was dated at yr .

440 yrs BP appears too young compared to the other dates below

(1150 yrs BP) and above (1120 yrs BP) it. Thus, it was also removed. 3.5. Quantiﬁcation of pollens and NPPs

The sample Beta-463577, which was dated to 1270 yrs BP, is

slightly older than the Beta-462835 and Poz-69617 samples that The pollen percentages are dominated by arboreal pollens (AP)

overlie it. Beta-463577 was probably reworked and has also been between 2100 and 1000 yrs cal. BP, with an average value of 80% of

removed from the age–depth model. The model was computed AP (Fig. 6-g). The percentages of non-arboreal pollens (NAP)

with Clam using a smooth relationship (Blaauw, 2010). increase from 1070 to 970 yrs cal. BP to reach an average value of

The age–depth model shows two major increase in the 50% between 970 yrs cal. BP and CE 1970. The NAP percentage

sedimentation rate, namely, one in the Middle Ages and one in reached a maximum of 65% at 200 yrs cal. BP and slightly decreased

the last century (Fig. 5-b). Unit 3 spans the period of 2100-1000 yrs during the last century (Fig. 6-g).

80 M. Bajard et al. / Anthropocene 24 (2018) 72–87

Fig. 6. Comparison of the geochemistry - LOI a), b) and C/N ratio c), total and available P analysis d), e) - of Lake Benit sediment core with total P in water (TPw) infered from

total carotenoids (TC) f), land-use evolution (pollen analysis g)), grazing activity (NPPs analysis h)) and the trophy of the lake (chironomid analysis i)). NAP = non-arboreal

pollen, AP = arboreal pollen. The pastoral/ruderal indicator curve includes Plantago lanceolata, P. major/media, P. alpina, Plantago sp., Urticaceae, Rumex acetosa/acetosella,

Artemisia, Chenopodiaceae, Centaurea cyanus, Papaver and is exaggerated x4. Dotted lines represent accumulation rates.

M. Bajard et al. / Anthropocene 24 (2018) 72–87 81

The concentration of strict coprophilous fungal ascospores is low pool, BENS5 is related to TiO2 and SiO2, and BENS3 and BENS4 are

À1

(under 12 spores. g )from2100to1040yrscal.BP(Fig. 6-h). It increases more closely related to the potassium-enriched part of the

at 1070 yrs cal. BP and reaches a maximum during the Middle Ages (66 terrigenous endmembers. The rocks from the limestone scree are

À1 À1

spores. g ). This concentration remains high (40 spores. g on related to CaO, and the rocks from the ﬂysch are closer to the

average) from 1070 to 200 yrs cal. BP, decreases after 200 yrs cal. BP potassium-enriched part of the terrigenous endmembers, between

and has remained below 20 during the last 50 years. The accumulation BENS3 and BENS4. The chemical signature of the soil transect sample

rates of fecal fungal spores are high during the Middle Ages (327 on the shore of the lake (A position) is the same as that of BENS5

-2 À1 -2 À1

spores. cm .yr ) and ranged otherwise from 0 to 94 spores. cm .yr , (Fig. 7-b). B position samples are more dispersed throughout the

reaching maximum values at 1300, 900 and 200 yrs cal. BP (Fig. 6-h). terrigenous endmembers. C position samples are close to the most

recent sediment samples, which are correlated to Dim 1. Sediments

3.6. Chironomid identiﬁcation between 2000 and 1000 yrs cal. BP are distributed between the CaO

endmembers and the siliceous part of the terrigenous endmembers

Among the 15 chironomid taxa identiﬁed in the sediment (Fig. 7-b). Sediments between1000 and 0 yrs cal. BP inage are mainly

samples, four of them represented more than 95% of the total attached to the potassium part of the terrigenous endmembers.

counts, namely, Tanytarsus lugens, Chironomus plumosus, Dicro-

tendipes sp. and Tanypodinae. T. lugens dominated the chironomid 4. Discussion

assemblages in association with Tanypodinae over most of the

studied time period with rather stable relative abundances (Fig. 6- 4.1. Origins of the sediments

i). A drastic change occurred ca CE 1950, when the relative

abundances of T. lugens and Tanypodinae decreased; these were 4.1.1. Mineral fraction

replaced by C. plumosus and Dicrotendipes, with C. plumosus The coarse carbonated fraction (from ﬁne sand to centimeter-

representing the new dominant taxa (Fig. 6-i). scale dropstones) of the sediment suggests that snow avalanches

from the limestone scree go down to the frozen lake in winter,

3.7. Mineral and organic geochemistry carrying carbonated materials (Fig. 4 and SI-II). The ice begins to

melt from the lake shores at the end of spring, creating ice patches

The C/N ratio ranges from 10 to 16.5 from 2100 to 1070 yrs cal. BP moving on the lake surface that then melt and deposit the material

and decreases until 900 yrs cal. BP (Fig. 6-c). It remains below 8 after at the bottom of the lake. The PCA highlights the two main sources

970 yrs cal. BP, except at 400 yrs cal. BP (14.4). The total P concentration of sediment which correspond to the limestone scree (from which

tends to slightly increase throughout the last 2100 years, with average the coarse carbonated fraction originates) and to the ﬂysch area

À1 À1

values of 613 mg.kg before 1000 yrs cal. BP and 766 mg.kg from (Fig. 7-b). The composition of the moraine ridge is heterogenous,

À1

1000 yrs cal. BP to CE 1950, and it reaches 1020 mg.kg at the top of mixing allochthone glacial deposits and autochthonous materials

the core (Fig. 6-d). The POlsen concentration ranges from 3.7 to 22.5 mg. from the limestone and the ﬂysch (Sesiano and Muller, 1986).

À1

kg over the last 2000 yrs (Fig. 6-e). The average concentration from Therefore, the corresponding parent material was not included in

À1

2000 to 900 yrs cal. BP is 12 mg.kg . It decreases after 900 yrs cal. BP the sampling strategy. However, this glacial formation is well

À1 À1

until reaching 3.7 mg.kg , except at 250 yrs cal. BP (14.7 mg.kg ). characterized by its soils (i.e., BENS5, and shore transects 1,2 and 3),

Then, the POlsen concentration increases from CE 1961 yrs to the as shown in the PCA (Fig. 7-b) and by the total P concentrations of

À1

present (15.4 mg.kg ). Theaccumulation rate of total carotenoids (TC) surface soils, higher in the ﬂysch area than they are in the moraine

is constant from 2100 to 1200 yrs cal. BP exhibits a mean value of ridge (Fig. 3-c). The sediments from 2000 to 1000 yrs cal. BP appear

À1 À1

0.02 mg.g .yr (Fig. 6-f). TC increases from 1200 yrs cal. BP, with a to have originated from the input of both limestone scree and soils

signiﬁcant peak between 1063 to 984 yrs cal. BP reaching upto 0.17 mg. that developed on the moraine ridge, while sediments of the last

À1 À1

g .yr . The accumulation rate then decreases and stabilizes in the millennia appear to be mainly derived from the ﬂysch area.

same order of magnitude as was observed before 1200 yrs cal. BP with Compared to the ﬂysch area, the soils of the moraine ridge seem to

À1 À1

an average value of 0.02 mg.g .yr . After 95 yrs cal. BP, it increases participate minorly in the mineral lake inputs, while its surface in

À1

again and reaches a new maximum in the last decade of 0.12 mg.g . the catchment is higher and highlights the more important

À1

yr . The inferred total P in water (TPw) displays the highest erodibility of soils that formed on the ﬂysch, over the last

À1

concentrations (10 to 15 mg.L ) and exhibits high variability between millennia, as reﬂected by the current gullies upstream. The most

2100 and 984 yrs cal. BP (Fig. 6-f). From 984 to 95 yrs cal. BP, the TPw important erodibility of the ﬂysch material also explains the glacial

À1

concentration remains stable, with an average value of 9.5 mg.L , and over-deepening that formed the lake in this region (Sesiano and

À1

it increases over the last 150 yrs up to 14 mg.L . Muller, 1986). The contribution of the ﬂysch to the sediments from

2000 to 1000 yrs cal. BP is not visible, perhaps due to the degree of

3.8. Geochemical endmembers weathering of soils before 1000 yrs cal. BP. Soils may have evolved

since the glacial retreat, i.e., approximately upper Dryas (11.8 ka cal.

The ﬁrst two dimensions of the PCA (Dim 1 and Dim 2) explain BP) and Preboreal (12.7 ka cal. BP) ages (Sesiano and Muller, 1986).

86% of the variability in the geochemical data of the sampled soils, It is thus possible that the long-term evolution of soils on the ﬂysch

rocks and sediments (Fig. 7). The PCA correlation circle shows three material led to soil with geochemical signatures close to those of

main endmembers (Fig. 7-a). The ﬁrst one is positively correlated to the moraine ridge. Then, they were disrupted and rejuvenated by

the ﬁrst component and yields high positive loadings for SiO2, TiO2, deforestation and agropastoral activities in the Middle-Ages,

Fe2O3, Al2O3, Al2O3/TiO2 and K2O, thus representing terrigenous increasing their potassium contents (Bajard et al., 2017a). The

materials. This endmember can be divided into two parts: one is K2O difference between BENS3 and BENS4 tends to conﬁrm this

positive on dimension 2, and the other is SiO2 and TiO2 negative on hypothesis. BENS4, which is located upstream of BENS3, is more

+ 2+

Dim 2. The other two endmembers, yielded by CaO and OM, weathered and depleted in exchangeable bases (i.e., K and Ca ), as

respectively, present negative loadings on Dim 1. CaO is positively reﬂected by its position in the PCA, where it is closer to BENS5 and

correlated to dimension 2, and OM is negatively correlated to SiO2 than K2O and CaO as BENS3 is (Fig. 7-a). Furthermore, we can

dimension 2. The sample map was colored based on the age of note the completely distinct geochemistry of the Umbrisol (BENS1)

sediment and the origins of the soil and rock samples (Fig. 7-b). The that developed on the stabilized limestone scree in the southern

soil proﬁles are well individualized. BENS1 is related to the organic region of the catchment (Figs. 2 and 7). Because of their highly

82 M. Bajard et al. / Anthropocene 24 (2018) 72–87

Fig. 7. Identiﬁcation of sediment sources to Lake Benit with PCA analysis including geochemistry of sediment core samples, soil samples (BENS soil proﬁles and A-B-C shore

transects) and rocks sampled in the limestone scree and in the ﬂysch. PCA allows the identiﬁcation of the sediment sources to Lake Bénit. a) Correlation circle. b) map of

individual samples colored by ages for sediment and origin of soil samples and rocks.

organic nature, these thick soils are rarely enhanced for lowest active layers, e.g., under coniferous forest (Duchaufour,

agricultural purposes and their contribution to erosion is low 1970). The interpretation of changes in the C/N ratio must be

(Bajard et al., 2017a). considered based on their different sources, i.e., terrestrial and

aquatic, as well as other proxies (e.g., erosion and land-use)

4.1.2. Organic matter because of possibly contrasting processes leading to these C/N

The variations in the C/N ratio are used to determine changes in values (Enters et al., 2006). In the Lake Bénit sediments, the low

the origin of OM and reﬂect local erosion processes and lacustrine OM accumulation rate associated with the to high C/N ratio is

productivity (Kaushal and Binford, 1999; Lerman et al., 1995). The interpreted to reﬂect an aquatic OM source, while a low C/N ratio

algal organic matter produced in lakes has lower C/N ratios (< 10) represents the input of terrestrial OM (Fig. 6-b-c).

than vascular plants (> 20) because vascular plants contain

cellulose and lignin (Meyers, 1994; Lerman et al., 1995; Tyson, 4.2. Evolution of the landscape and human activities of Lake Bénit over

2012). Among vascular plants, arboreal woody plants that have the last 2100 yrs

more lignin than herbs also have higher C/N ratios (e.g., 40–45 for

beech leaf, 65 for conifers). However, their degradation leads to 4.2.1. From 2100 to 1100 yrs cal. BP (150 BCE to CE 850): an

lower C/N ratios (e.g., litter), which is also true for aquatic plants undisturbed forested system

(Tyson, 2012). Thus, the C/N ratios in soils range from 8 in the most From 2100 to 110 0 yrs cal. BP, erosion was low and the

active layer (e.g., in meadows or active forest litters) to 40 in the proportion of tree pollens (Fig. 6-a-g), mainly spruce, alder and

M. Bajard et al. / Anthropocene 24 (2018) 72–87 83

beech, was high, thus indicating a forested area. The high C/N ratio could reﬂect a decrease in the local pastoral pressure, as is also

indicates inputs of arboreal organic matter and conﬁrms that a suggested by the decrease in the proportion of ruderal pollen

forested landscape existed around the lake (Figs. 6-c and 8 -a). during the late Middle Ages. The consistent nature of inputs whit

Although terrigenous inputs to the lake came from the limestone decreasing grazing activities highlights the deep modiﬁcations of

scree during avalanches and from the runoff on the slopes around the catchment. There is no resilience, and this new steady state

the lake, they mainly came from the moraine area that covers the could result from the progressive stabilization of the pastoral area

main part of the lakeshore (Fig. 7). The very low abundances of following the erosive crisis triggered by the deforestation.

coprophilous fungal spores and pastoral pollen indicators indicate From 300 to 150 yrs cal. BP (CE 1650–1800), the available data

the lack of or low grazing activity in the catchment, which is concerning P concentrations, NAP and ruderal pollens percentages,

unusual for a subalpine area in this region and during this period. and strict coprophilous fungal spores indicate that pastoral

The main deforestation activities for livestock grazing began activities increase again during the Little Ice Age (Fig. 6-D-e-g-h).

during the Neolithic (ca. 7500-4500 BP) and increased during the Over the entire period between 900 yrs cal. BP and CE 1950, the

Bronze Age (ca. 4500-3000 BP) and the Roman period (ca. 2500- dominant abundances of chironomid T. lugens and Tanypodinae

1500 BP) (Giguet-Covex et al., 2014; Valese et al., 2014; Bajard et al., remained relatively steady (Fig. 6-i). T. lugens and Tanypodinae are

2016; Pini et al., 2017). The geographic impossibility of moving mainly associated with well oxygenated waters (Saether, 1979;

cattle up to pastures higher than 1600 m to follow the grass growth Hirabayashi et al., 2004; Brooks et al., 2007; Millet et al., 2010) and

may not have increased interest in this pasture area during the could indicate that the lake existed in an oligotrophic state until

Roman period, while other surrounding mountains offered the the 1950’s. But due to the use of a non-standard mesh size, these

possibility for pasture due to their longer exposure to sun. During results could be biaised. These observations are nonetheless

this period, the concentration of total P in water inferred from the consistent with both the concentration of total P in water inferred

total carotenoid concentrations suggests the lake was in a from TC, which also indicates the oligotrophic state of the lake

mesotrophic state. (Nürnberg, 1996; Wetzel, 2001) between 1000 and 200 yrs cal. BP

(Fig. 6-f) and with the constant concentration of total P in the

4.2.2. From 1100 to 0 yrs cal. BP (CE 850–1950): opening of the sediment and the slight decrease of available P, except during the

environment and development of grazing activities period of 300 - 150 yrs cal. BP (Fig. 6-d-e). However, the total P in

Between 110 0 and 1000 yrs cal. BP, erosion increased from 10 to water inferred from TC increases gradually from CE 1900 while no

À2 -1

50 mg.cm .yr and reached its maximum throughout the last 2000 substantial change in the chironomid community were recorded

years (Fig. 6-a). During the 9 century, the proportion of AP before CE 1950 with our non-standard method.

percentages decreased sharply to the beneﬁt of ruderal plants,

suggesting a quick change in the composition of local vegetation, 4.2.3. Consequences of human-induced modiﬁcations in the 20

which is supported by the decrease in the C/N ratio. The catchment century

was deforested and used as a pasture area with the development of After the 1950’s, all of the proxies indicate a major environmental

grazing activities, as suggested by both the strictcoprophilous fungal change (Fig. 6). Both the concentrations and accumulation rates of

ascospore concentrations and accumulation rates (Figs. 6-h and 8 - organic matter increase (Fig. 6-a-b). The C/N ratio increases slightly

b). The late development of human activities in the Lake Benit to 8, indicating inputs from the watershed (Fig. 6-c). The percentages

catchment may be due to the increase in demography in the Middle of NAP, including those of ruderal pollen, and coprophilous fungal

Ages in the French Alps. This increase would have trigger higher ascospores, decrease, indicating the reduction of local pastoral

needs in pasture and wood to sustain the development of societies. activities (Fig. 6-g-h), as conﬁrmed by the recovery of spruce. The

As indicated by the PCA (Fig. 7), the mineral inputs changed, with concentrations of both total and available P in the sediment also

the disappearance of the coarse carbonated fraction from the increase. The concentration of total P in the water reaches a

limestone scree and the destabilization of the ﬂysch area due to maximum, and the chironomid assemblages drastically change

grazing. The causes of the lower carbonate inputs are not completely (Fig. 6-i). The dominance of C. plumosus is indicative of oxygen

understood. It ﬁrst suggests a decrease (or a stop) of the avalanche depletion conditions at the bottom of the lake. Eutrophication may

episodes that cannot be explained by climate variability since the be a primary driver of the change in oxygen conditions as suggested

Middle Ages. There is no variability in the carbonate content by other proxies. Changes in ice coverage duration can also foster

corresponding to a warmer climate in the medieval period and to a hypoxic conditions and inﬂuence chironomid assemblages (Gran-

colder climate in the following Little Ice Age. Moreover, the ados and Toro, 2000). Nonetheless, the relative increase in annual air

deforestation of the catchment would have led to an increase in temperature over the recent decades does not support a major

avalanches, unless the deforestation led to other avalanche implicationof the durationoficecoverage durationindrivingoxygen

conditions, such as smaller but much more regular avalanches that dynamic. Dicrotendipes is associated with littoral zones and with

brought fewer coarse materials into the frozen lake. The stabilization macrophytes (Engels and Cwynar, 2011). Changes in the chironomid

of the scree is also possible, but its synchronization with the change assemblage could therefore result from the eutrophication of the

in landscape remains obscure. The more important contribution of lake. The evolution of the lake trophic status could be associated

the ﬂysch area after 1000 yrs cal. BP is explained by the increased with the high concentrations of P in the water reached in the 1950’s,

erodibility of thisformation to runoff compared tothe moraine ridge, which are reﬂected in the sediment by both the total and available P

which was emphasized by grazing and easily formed gullies on the concentrations. This excess nutrient reﬂected by P concentrations

deepest slopes. Furthermore, the position of the ﬂysch formation on could have triggered both the eutrophication and the increase in the

a ﬂat area represents an ideal place for cattle grazing and could have OM concentration of the sediment.

accentuated erosion.

Erosion decreases after 1000 yrs cal. BP during the Middle Ages 4.3. Recent eutrophication associated with past nutrient storage and

and stabilizes after 700 yrs cal. BP during the late Middle Ages to a recent lake-level increase

higher level than was observed before 110 0 yrs cal. BP (Fig. 6-a).

The nature of the inputs is also stabilized until CE 1900, as shown 4.3.1. Development of new activities and eutrophication process

by the constant concentrations of most of the geochemical proxies dynamics

(i.e., LOI, C/N ratio, total and available P). However, the accumula- Following the reduction of pastoral activities, new leisure

tion rates of the coprophilous fungal ascospores decrease, which activities developed around the lake, such as hiking and ﬁshing.

84 M. Bajard et al. / Anthropocene 24 (2018) 72–87

Fish introductions were reported from at least the end of the 19 mainly grazed on the lower ﬂatter parts of the pasture (Jewell et al.,

century (Druart et al., 1999) and could explain the rise in the 2007), i.e., on the shore of the lake, close to the water. Animals

concentration of P in water that occurred at that time, prior to the would have concentrated their dungs, and thus accumulated

chironomid community changes. With the rise of hiking and nutrients, especially P, in these areas where they would likely have

ﬁshing, a summer bar was built on the shore of the lake in CE 1952 rested and ruminated (Fig. 8-c). It is even possible that these main

and the lake level was raised in CE 1964 to flood the macrophyte flat areas may have been flooded in CE 1964. Moreover, the

(Equisetum sp.) belt surrounding the lake to allow easier access for trampling on the shores induced by the concentration of cattle in

anglers. The raising of the lake outlet in CE 1964 triggered a lake these patches could have strengthened their destabilization after

level rise of 2 m and doubled the lake area. The lake level ﬂuctuated ﬂooding and increased the release of P (Marlow and Pogacnik,

for several years before stabilizing, triggering bank undercutting 1985; Bennett et al., 2001).

and ﬂoating mottles. The modiﬁcation and concentration of nutrients such as P in

Flooded soils currently contain less total and available P soils can last long after their input (Reed-Andersen et al., 2000;

compared to the non-ﬂooded soils surrounding the lake, especially Dupouey et al., 2002; Bennett et al., 2001). The effect of grazing,

those on the moraine ridge shore, suggesting a release of P from since the Middle Ages could have resulted in the P enrichment of

ﬂooded soils with the lake-level increase (Fig. 3). However, the the soil on the shores of Lake Bénit. Therefore, we suggest that the

distribution of available P on the ﬂysch area (transect 4, Fig. 3-b) combined effects of historical grazing and the recent lake-level

presents the opposite trend, with more important contents increase for ﬁsheries is responsible for recent eutrophication of the

observed in the ﬂooded soil (position C) than on the shore lake (Fig. 8-d).

(position A). This effect can be explained by the higher erodibility

of the ﬂysch: the available P is mainly contained in surface soil 4.3.3. Aftereffect of the diachronous signals of anthropogenic effects

(Fig. 3-a-A) and accumulates downstream with erosion. The same The recent eutrophication of Lake Bénit highlights the

process can also explain the consistent concentrations of total P in combination of anthropogenic effects at different time scales.

the ﬂysch area (Fig. 3-b, i.e., transect 4, BENS3 and BENS4). The Medieval grazing practices have had direct effects on the

steady erosion in this area appears to homogenize the total P in the environment, such as the modiﬁcations of landscape and erosion,

surface soil. Thus, the increases in total and available P, which could thereby triggering after-effects of the long soil storage of nutrients,

be responsible for the lake eutrophication, are likely due to the such as the recent lake eutrophication, which behaves partly as a

release of soil from the moraine ridge following the lake-level “palaeo-anthropocene” effect. The recent release of soil P with

increase. Hikers and the settlement of the summer bar could not lake-level increase derived from long-standing practices ﬁrst

have triggered such an abrupt effect on the nutrient supply, while underlines the strong ability of soils to buffer the anthropic

grazing had already lasted for several centuries without causing a impacts on freshwater ecosystems and their ability to become

noticeable evolution in the chironomid community. Moreover, sources when environmental conditions change. A similar

several studies have observed eutrophication following lake-level feedback process was shown by Sabatier et al. (2014) in a vineyard

ﬂuctuations associated with P excess (Hambright et al., 2004; with the release of banned remnant pesticide (DDT) in a lake after

Punning et al., 2008; Moos et al., 2009), which supports this recent herbicide-induced erosion. The remobilization of P also

hypothesis. underlines the diachronous onset of the “Anthropocene”. Past

human activities are not necessary immediately visible but can

4.3.2. Modiﬁcation of P cycling with human activities appear and last long afterwards (Aubert, 1960; Edgeworth et al.,

The higher P values in the surface soil of the shore than 2015), as a result of new practices and new planning, thus

downward could have a natural and/or anthropic origin. Phospho- intensifying their impacts. This result strongly suggests consider-

rus is naturally enriched in top soil, as it is recycled by vegetation, ing old human practices when establishing management policies

and higher P concentrations are often reported in pastured and of territories in order to carefully adapt to new needs. Furthermore,

grassland areas compared to forested areas (Ross et al., 1999; the after-effects, i.e., P release, are recycled and emphasized by the

Jobbágy and Jackson, 2001; Wang et al., 2009). Between 2000 and action of introduced ﬁshes. Fishes can continuously suspend

1000 yrs cal. BP, the mean total P concentration in the sediments phosphorus that is ﬁxed in surface lake sediments, i.e., particulate

À1

was lower (613 mg.kg ) but slowly increased, relative to that P, or that in undecomposed organic matter (Elizabeth et al., 1992;

observed after deforestation under grassland (between 1000 yrs Huser et al., 2016), and also by increasing human touristic pressure

À1

cal. BP and CE 1950; 766 mg.kg ). After the rise in lake level, the (e.g., Druart et al., 1999; Hadwen et al., 2003; Ostendorp et al.,

À1

concentration reached 1020 mg.kg . The exports of total P from 2004; Dokulil, 2014). The multiplication and superposition of

the watershed tend to be higher under grassland than under forest human activities are a chain of events working as a cascade effect.

and depend on the geological substrate (Dillon and Kirchner, 1975; They accentuated the human imprint on the environment in the

Dorioz and Trevisan, 2013). An increase in the total P concentration last century and both participate and sustain the idea of the “great

of sediment from the Middle Ages could have resulted from both acceleration” (Steffen et al., 2015).

the modiﬁcations in land use and associated changes in the

sediment sources within the watershed. However, the available P 5. Conclusion

source is more important on the moraine ridge shore (Fig. 3-b) and

will thus contribute more actively to eutrophication, while it only The combined analyses of geochemistry (including major

slightly affects to particulate inputs. elements, C, N and P), pollen, coprophilous fungal ascospores

Herbivores play a major role in nutrient cycling (e.g., Williams and chironomids of the Lake Bénit sediment sequence allowed us

and Haynes, 1990; Sitters et al., 2017). Higher P concentration in to reconstruct past dynamics of the catchment and changes in the

mountain pastured surface soils can result in two processes, trophic lake status. Performing a comparison with soil geochem-

namely, anthropic inputs, such as the spreading of organic istry allowed us to determine the origin of sediment in the

fertilization, and the natural distribution of animal feces, which catchment and to understand the relationship between practices

is concentrated in the remaining and ruminate areas. In the (e.g., agriculture, leisure) and observed processes (e.g., erosion,

extensive pasture context of the Lake Bénit catchment, spreading is eutrophication). The Lake Bénit sediment sequence spans the last

not practiced. However, the redistribution and concentration of 2100 years. The catchment was forested until the Middle Ages,

animal excrement close to the lake is strongly conceivable, as cattle without extensive human activities. Sediments came from

M. Bajard et al. / Anthropocene 24 (2018) 72–87 85

the soils on the shore of the lake, which were analyzed along four

transects from the ﬂooded area to the out of-water area, suggest

that a release of P from the ﬂooded soils was responsible for the

eutrophication. Grazing animals play a major role in nutrient

cycling and the long-grazing practices of the last millennia would

have led to the high P concentration on the shores of the lake. Our

study highlights the combined effects of past and recent activities

on the current eutrophication process, as well as the legacy of early

anthropogenic effects through soil destabilization that must be

more taken into account for the future management of social ecosystems.

Data References

Sedimentological and geochemical data (i.e., XRF, LOI, Grain-

size, CNP and total P analyses) will be available on https://www. pangaea.de/.

Acknowledgements

We ﬁrst thank Michel Dorioz who brought us many times to the

lake for coring and soil sampling, and Mr and Mrs Bastian, who

have owned the Lake Bénit refreshment bar since 1957 and

provided us with historical information about the past activities

around the lake and their observations of the modiﬁcations of the

environment. We also thank the villages of Marnaz and Mont-

Saxonnex for sampling authorization and support from the ski

resort for winter access to the lake. The authors thank Annie

Millery for the analyses of photosynthetic pigments and University

Savoie Mont Blanc for funding (PRETL Project). We also thank the

Laboratoire Souterrain de Modane facilities for the gamma

spectrometry measurements. Lastly, we thank Veerle Vanacker

and two anonymous reviewers for their comments and sugges-

tions that improved the manuscript.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in

the online version, at doi:https://doi.org/10.1016/j. ancene.2018.11.005.

References

Adrian, R., O’Reilly, C.M., Zagarese, H., Baines, S.B., Hessen, D.O., Keller, W.,

Livingstone, D.M., Sommaruga, R., Straile, D., Van Donk, E., Weyhenmeyer, G.A.,

Winder, M., 2009. Lakes as sentinels of climate change. Limnol. Oceanogr. 54,

2283–2297. doi:http://dx.doi.org/10.4319/lo.2009.54.6_part_2.2283.

Appleby, P., Oldﬁeld, F., 1978. The calculation of lead-210 dates assuming a constant

rate of supply of unsupported 210 Pb to the sediment. Catena 5, 1–8.

Appleby, P., Richardson, N., Nolan, P., 1991. 241Am dating of lake sediments.

Environmental History and Palaeolimnology. Springer, pp. 35–42.

Arnaud, F., Poulenard, J., Giguet-Covex, C., Wilhelm, B., Révillon, S., Jenny, J.-P., Revel,

M., Enters, D., Bajard, M., Fouinat, L., others, 2016. Erosion under climate and

human pressures: An alpine lake sediment perspective. Quat. Sci. Rev.152,1–18.

Arnaud, F., Révillon, S., Debret, M., Revel, M., Chapron, E., Jacob, J., Giguet-Covex, C.,

Fig. 8. Evolution of the Lake Benit catchment and heritage of early anthropic Poulenard, J., Magny, M., 2012. Lake Bourget regional erosion patterns

activities on the current functionning of the lake-catchment system with reconstruction reveals Holocene NW European Alps soil evolution and

paleohydrology. Quat. Sci. Rev. 51, 81–92. doi:http://dx.doi.org/10.1016/j.

eutrophication triggered by P release from soils after 1960’s lake level rise.

quascirev.2012.07.025.

Aubert, G., 1960. Classiﬁcation Et Cartographie Des Sols : Observations à La Suite Du

7ème Congrès International De La Science Du Sol. Congrès International De La

avalanches of limestone scree and runoff of the moraine ridge with

Science Du Sol. 7., s.l. (US). , pp. 901–903.

very low inputs. After 110 0 yrs cal. BP, the catchment was

Baize, D., 2000. Guide des analyses en pédologie: 2e édition, revue et augmentée.

deforested for cattle grazing. Erosion increased and mainly from Editions Quae. .

the ﬂysch area that was destabilized and gullied. Chironomid Bajard, M., Poulenard, J., Sabatier, P., Develle, A.-L., Giguet-Covex, C., Jacob, J.,

Crouzet, C., David, F., Pignol, C., Arnaud, F., 2017a. Progressive and regressive soil

assemblages indicate that the lake was oligotrophic. The changes in

–

th evolution phases in the Anthropocene. CATENA 150, 39 52. doi:http://dx.doi.

assemblages in the second half of the 20 century underline the org/10.1016/j.catena.2016.11.001.

still-visible eutrophication of the lake. Eutrophication is associated Bajard, M., Poulenard, J., Sabatier, P., Etienne, D., Ficetola, F., Chen, W., Gielly, L.,

Taberlet, P., Develle, A.-L., Rey, P.-J., Moulin, B., Beaulieu, J.-L.de, Arnaud, F.,

with the change in activities in the catchment, i.e., ﬁshing and

2017b. Long-term changes in alpine pedogenetic processes: Effect of millennial

hiking, leading to a lake-level increase in CE 1964, while pastoral

agro-pastoralism activities (French-Italian Alps). Geoderma 306, 217–236. doi:

activity decreased. The available and total phosphorus contents in http://dx.doi.org/10.1016/j.geoderma.2017.07.005.

86 M. Bajard et al. / Anthropocene 24 (2018) 72–87

Bajard, M., Sabatier, P., David, F., Develle, A.-L., Reyss, J.-L., Fanget, B., Malet, E., Germany. Sci. Total Environ. 369, 178–187. doi:http://dx.doi.org/10.1016/j.

Arnaud, D., Augustin, L., Crouzet, C., Poulenard, J., Arnaud, F., 2016. Erosion scitotenv.2006.05.020.

record in Lake La Thuile sediments (Prealps, France): evidence of montane Etienne, D., Jouffroy-Bapicot, I., 2014. Optimal counting limit for fungal spore

landscape dynamics throughout the Holocene. Holocene 26, 350–364. abundance estimation using Sporormiella as a case study. Veg. Hist. Archaeobot.

Bennett, E.M., Carpenter, S.R., Caraco, N.F., 2001. Human Impact on Erodable 23, 743–749. doi:http://dx.doi.org/10.1007/s00334-014-0439-1.

phosphorus and eutrophication: a global perspective: increasing accumulation Etienne, D., Wilhelm, B., Sabatier, P., Reyss, J.-L., Arnaud, F., 2013. Inﬂuence of sample

of phosphorus in soil threatens rivers, lakes, and coastal oceans with location and livestock numbers on Sporormiella concentrations and

eutrophication. BioScience 51, 227–234. accumulation rates in surface sediments of Lake Allos, French Alps. J.

Beug, H.-J., 2004. Leitfaden der Pollenbestimmung für Mitteleuropa und Paleolimnol. 49, 117–127. doi:http://dx.doi.org/10.1007/s10933-012-9646-x.

angrenzende Gebiete.. . Faegri, K., Iversen, J., 1989. Textbook of Pollen Analysis, revised by Faegri K, Kaland

Blaauw, M., 2010. Methods and code for ‘classical’ age-modelling of radiocarbon PE, Krzywinski K. J. Wiley, New York.

sequences. Quat. Geochronol. 5, 512–518. doi:http://dx.doi.org/10.1016/j. FAO, 2006. Guidelines for soil description, 4., [rev.] ed. ed. FAO, Rome.

quageo.2010.01.002. Giguet-Covex, C., Arnaud, F., Poulenard, J., Disnar, J.-R., Delhon, C., Francus, P., David,

Brisset, E., Guiter, F., Miramont, C., Delhon, C., Arnaud, F., Disnar, J.-R., Poulenard, J., F., Enters, D., Rey, P.-J., Delannoy, J.-J., 2011. Changes in erosion patterns during

Anthony, E., Meunier, J.-D., Wilhelm, B., Pailles, C., 2012. Approche the Holocene in a currently treeless subalpine catchment inferred from lake

multidisciplinaire d’une séquence lacustre holocène dans les Alpes du Sud au sediment geochemistry (Lake Anterne, 2063 m a.s.l., NW French Alps): the role

lac Petit (Mercantour, alt.2200m. France): histoire d’un géosystème dégradé. of climate and human activities. Holocene 21, 651–665. doi:http://dx.doi.org/

Quaternaire. , pp. 23. 10.1177/0959683610391320.

Brisset, E., Miramont, C., Guiter, F., Anthony, E.J., Tachikawa, K., Poulenard, J., Arnaud, Giguet-Covex, C., Pansu, J., Arnaud, F., Rey, P.-J., Griggo, C., Gielly, L., Domaizon, I.,

F., Delhon, C., Meunier, J.-D., Bard, E., Sumera, F., 2013. Non-reversible geosystem Coissac, E., David, F., Choler, P., Poulenard, J., Taberlet, P., 2014. Long livestock

destabilisation at 4200 cal. BP: sedimentological, geochemical and botanical farming history and human landscape shaping revealed by lake sediment DNA.

markers of soil erosion recorded in a Mediterranean alpine lake. Holocene 23, Nat. Commun. 5 doi:http://dx.doi.org/10.1038/ncomms4211.

1863–1874. doi:http://dx.doi.org/10.1177/0959683613508158. Goldberg, E.D., 1963. Geochronology with 210Pb. Radioactive Dating 121–131.

Brodersen, K.P., Quinlan, R., 2006. Midges as palaeoindicators of lake productivity, Granados, I., Toro, M., 2000. Recent warming in a high mountain lake (Laguna

eutrophication and hypolimnetic oxygen. Quat. Sci. Rev. 25, 1995–2012. doi: Cimera, Central Spain) inferred by means of fossil chironomids. J. Limnol. 59,

http://dx.doi.org/10.1016/j.quascirev.2005.03.020. 109–119.

Brooks, S.J., Langdon, P.G., Heiri, O., 2007. The Identiﬁcation and Use of Palaearctic Guilizzoni, P., Marchetto, A., Lami, A., Gerli, S., Musazzi, S., 2011. Use of sedimentary

Chironomidae larvae in Palaeoecology. Quaternary Research Association. pigments to infer past phosphorus concentration in lakes. J. Paleolimnol. 45,

Carpenter, S.R., Benson, B.J., Biggs, R., Chipman, J.W., Foley, J.A., Golding, S.A., 433–445. doi:http://dx.doi.org/10.1007/s10933-010-9421-9.

Hammer, R.B., Hanson, P.C., Johnson, P.T.J., Kamarainen, A.M., Kratz, T.K., Hadwen, W.L., Arthington, A.H., Mosisch, T.D., 2003. The impact of tourism on dune

Lathrop, R.C., McMahon, K.D., Provencher, B., Rusak, J.A., Solomon, C.T., Stanley, lakes on Fraser Island, Australia. Lakes Reserv.: Res. Manage. 8, 15–26.

E.H., Turner, M.G., Vander Zanden, M.J., Wu, C.-H., Yuan, H., 2007. Understanding Hambright, K.D., Eckert, W., Leavitt, P.R., Schelske, C.L., 2004. Effects of historical

regional change: a comparison of two lake districts. BioScience 57, 323–335. lake level and land use on sediment and phosphorus accumulation rates in Lake

doi:http://dx.doi.org/10.1641/B570407. Kinneret. Environ. Sci. Technol. 38, 6460–6467. doi:http://dx.doi.org/10.1021/

Carpenter, S.R., Christensen, D.L., Cole, J.J., Cottingham, K.L., He, X., Hodgson, J.R., es0492992.

Kitchell, J.F., Knight, S.E., Pace, M.L., et al., 1995. Biological control of Heiri, O., Lotter, A.F., Lemcke, G., 2001. Loss on ignition as a method for estimating

eutrophication in lakes. Environ. Sci. Technol. 29, 784–786. doi:http://dx.doi. organic and carbonate content in sediments: reproducibility and comparability

org/10.1021/es00003a028. of results. J. Paleolimnol. 25, 101–110.

Coops, H., Beklioglu, M., Crisman, T.L., 2003. The role of water-level ﬂuctuations in Hirabayashi, K., Yoshizawa, K., Yoshida, N., Kazama, F., 2004. Progress of

shallow lake ecosystems–workshop conclusions. Hydrobiologia 506, 23–27. eutrophication and change of chironomid fauna in Lake Yamanakako. Jpn. J.

Correll, D.L., 1998. The role of phosphorus in the eutrophication of receiving waters: Limnol. 5, 47–53. doi:http://dx.doi.org/10.1007/s10201-003-0113-2.

a review. J. Environ. Qual. 27, 261–266. Huser, B.J., Bajer, P.G., Chizinski, C.J., Sorensen, P.W., 2016. Effects of common carp

Costanza, R., Graumlich, L., Steffen, W., Crumley, C., Dearing, J., Hibbard, K., Leemans, (Cyprinus carpio) on sediment mixing depth and mobile phosphorus mass in

R., Redman, C., Schimel, D., 2007. Sustainability or collapse: what can we learn the active sediment layer of a shallow lake. Hydrobiologia 763, 23–33. doi:

from integrating the history of humans and the rest of nature? Ambio A J. Hum. http://dx.doi.org/10.1007/s10750-015-2356-4.

Environ. 36, 522–527. Jenny, J.-P., Arnaud, F., Dorioz, J.-M., Covex, C.G., Frossard, V., Sabatier, P., Millet, L.,

Davidson, T.A., Jeppesen, E., 2013. The role of palaeolimnology in assessing Reyss, J.-L., Tachikawa, K., Bard, E., others, 2013. A spatiotemporal investigation

eutrophication and its impact on lakes. J. Paleolimnol. 49, 391–410. doi:http:// of varved sediments highlights the dynamics of hypolimnetic hypoxia in a large

dx.doi.org/10.1007/s10933-012-9651-0. hard-water lake over the last 150 years. Limnol. Oceanogr. 58, 1395–1408.

Dillon, P.J., Kirchner, W.B., 1975. The effects of geology and land use on the export of Jewell, P.L., Käuferle, D., Güsewell, S., Berry, N.R., Kreuzer, M., Edwards, P.J., 2007.

phosphorus from watersheds. Water Res. 9, 135–148. Redistribution of phosphorus by cattle on a traditional mountain pasture in the

Dokulil, M.T., 2014. Environmental Impacts of Tourism on Lakes, Ansari, A.A., Gill, S. Alps. Agric. Ecosyst. Environ. 122, 377–386. doi:http://dx.doi.org/10.1016/j.

S. Eutrophication: Causes, Consequences and Control. Springer Netherlands, agee.2007.02.012.

Dordrecht, pp. 81–88. https://https://doi.org/doi.org/10.1007/978-94-007- Jobbágy, E.G., Jackson, R.B., 2001. The distribution of soil nutrients with depth:

7814-6_7. global patterns and the imprint of plants. Biogeochemistry 53, 51–77.

Dorioz, J.-M., Trevisan, D., 2013. Le transfert diffus du phosphore dans les bassins Jordan-Meille, L., Dorioz, J.-M., Wang, D., 1998. Analysis of the export of diffuse

agricoles: ordres de grandeur, mécanismes, maîtrise. Revue EAT thématique 27–47. phosphorus from a small rural watershed. Agronomie 18, 5–26. doi:http://dx.

Doyen, E., Etienne, D., 2017. Ecological and human land-use indicator value of fungal doi.org/10.1051/agro:19980101.

spore morphotypes and assemblages. Veg. Hist. Archaeobot. 26, 357–367. doi: Kaushal, S., Binford, M.W., 1999. Relationship between C: N ratios of lake sediments,

http://dx.doi.org/10.1007/s00334-016-0599-2. organic matter sources, and historical deforestation in Lake Pleasant,

Druart, J.-C., Dorioz, J.-M., Blanc, P., Balvay, G., 1999. Un lac à protéger: le lac Bénit Massachusetts, USA. J. Paleolimnol. 22, 439–442.

(Haute-Savoie). Revue de géographie alpine 87, 87–99. Krishnaswamy, S., Lal, D., Martin, J.M., Meybeck, M., 1971. Geochronology of lake

Dubois, N., Saulnier-Talbot, É., Mills, K., Gell, P., Battarbee, R., Bennion, H., Chawchai, sediments. Earth Planet. Sci. Lett. 11, 407–414. doi:http://dx.doi.org/10.1016/

S., Dong, X., Francus, P., Flower, R., Gomes, D.F., Gregory-Eaves, I., Humane, S., 0012-821X(71)90202-0.

Kattel, G., Jenny, J., Langdon, P., Massaferro, J., McGowan, S., Mikomägi, A., Ngoc, Langdon, P.G., Ruiz, Z., Wynne, S., Sayer, C.D., Davidson, T.A., 2010. Ecological

N.T.M., Ratnayake, A.S., Reid, M., Rose, N., Saros, J., Schillereff, D., Tolotti, M., inﬂuences on larval chironomid communities in shallow lakes: implications for

Valero-Garcés, B., 2018. First human impacts and responses of aquatic systems: palaeolimnological interpretations. Freshw. Biol. 55, 531–545. doi:http://dx.doi.

a review of palaeolimnological records from around the world. Anthr. Rev. 5, org/10.1111/j.1365-2427.2009.02345.x.

28–68. doi:http://dx.doi.org/10.1177/2053019617740365. Lerman, A., Imboden, D.M., Gat, J.R. (Eds.), 1995. Physics and Chemistry of Lakes..

Dudgeon, D., Arthington, A.H., Gessner, M.O., Kawabata, Z.-I., Knowler, D.J., Lévêque, Springer, Berlin Heidelberg, Berlin, Heidelberg doi:http://dx.doi.org/10.1007/

C., Naiman, R.J., Prieur-Richard, A.-H., Soto, D., Stiassny, M.L.J., Sullivan, C.A., 978-3-642-85132-2.

2006. Freshwater biodiversity: importance, threats, status and conservation Marlow, C.B., Pogacnik, T.M., 1985. Time of grazing and cattle-induced damage to

challenges. Biol. Rev. 81, 163. doi:http://dx.doi.org/10.1017/ streambanks. RR Johnson, CD Zeibell, DR Patton, PF Ffolliot, and RH Hamre

S1464793105006950. (Technical Coordinators) Riparian ecosystems and their management:

Dupouey, J.-L., Dambrine, E., Lafﬁte, J.-D., Moares, C., 2002. Irreversible impact of reconciling conﬂicting uses. GTR RM-120. USDA Forest Service, Rocky Mountain

past land use on forest soils and biodiversity. Ecology 83, 2978–2984. Forest and Range Experimental Station, Fort Collins, Colorado, pp. 279–284.

Edgeworth, M., deB Richter, D., Waters, C., Haff, P., Neal, C., Price, S.J., 2015. Meyers, P.A., 1994. Preservation of elemental and isotopic source identiﬁcation of

Diachronous beginnings of the Anthropocene: the lower bounding surface of sedimentary organic matter. Chem. Geol. 114, 289–302.

anthropogenic deposits. Anthr. Rev. 2, 33–58. Millet, L., Giguet-Covex, C., Verneaux, V., Druart, J.-C., Adatte, T., Arnaud, F., 2010.

Elizabeth, K., Getachew, T., Taylor, W., Zinabu, G.-M., 1992. Eutrophication of Lake Reconstruction of the recent history of a large deep prealpine lake (Lake

Hayq in the Ethiopian highlands. J. Plankton Res. 14, 1473–1482. Bourget, France) using subfossil chironomids, diatoms, and organic matter

Engels, S., Cwynar, L.C., 2011. Changes in fossil chironomid remains along a depth analysis: towards the deﬁnition of a lake-speciﬁc reference state. J. Paleolimnol.

gradient: evidence for common faunal thresholds within lakes. Hydrobiologia 44, 963–978. doi:http://dx.doi.org/10.1007/s10933-010-9467-8.

665, 15–38. doi:http://dx.doi.org/10.1007/s10750-011-0601-z. Moos, M.T., Laird, K.R., Cumming, B.F., 2009. Climate-related eutrophication of a

Enters, D., Lücke, A., Zolitschka, B., 2006. Effects of land-use change on deposition small boreal lake in northwestern Ontario: a palaeolimnological perspective.

and composition of organic matter in Frickenhauser See, northern Bavaria, Holocene 19, 359–367. doi:http://dx.doi.org/10.1177/0959683608101387.

M. Bajard et al. / Anthropocene 24 (2018) 72–87 87

Mourier, B., Poulenard, J., Carcaillet, C., Williamson, D., 2010. Soil evolution and Schindler, D.W., 2012. The dilemma of controlling cultural eutrophication of lakes.

subalpine ecosystem changes in the French Alps inferred from geochemical Proc. R. Soc. B: Biol. Sci. 279, 4322–4333. doi:http://dx.doi.org/10.1098/

analysis of lacustrine sediments. J. Paleolimnol. 44, 571–587. doi:http://dx.doi. rspb.2012.1032.

org/10.1007/s10933-010-9438-0. Schindler, D.W., 2009. Lakes as sentinels and integrators for the effects of climate

Murphy, J., Riley, J.P., 1962. A modiﬁed single solution method for the determination change on watersheds, airsheds, and landscapes. Limnol. Oceanogr. 54, 2349–

of phosphate in natural waters. Anal. Chim. Acta 27, 31–36. 2358. doi:http://dx.doi.org/10.4319/lo.2009.54.6_part_2.2349.

Nürnberg, G.K.,1996. Trophic state of clearand colored, soft- and hardwater lakes with Schindler, D.W., Kidd, K.A., Muir, D.C.G., Lockhart, W.L., 1995. The effects of

special consideration of nutrients, anoxia, phytoplankton and ﬁsh. Lake Reserv. ecosystem characteristics on contaminant distribution in northern freshwater

Manage. 12, 432–447. doi:http://dx.doi.org/10.1080/07438149609354283. lakes. Sci. Total Environ. 160–161, 1–17. doi:http://dx.doi.org/10.1016/0048-

Olsen, S.R., 1954. Estimation of Available Phosphorus in Soils by Extraction With 9697(95)04340-7.

Sodium Bicarbonate. United States Department Of Agriculture, Washington. Sesiano, J., 1993. Monographie Physique Des Plans d’eau Naturels Du Département

Ostendorp, W., Schmieder, K., Jöhnk, K.D., 2004. Assessment of human pressures De La Haute-savoie, France. .

and their hydromorphological impacts on lakeshores in Europe. Int. J. Sesiano, J., Muller, A., 1986. Quelques observations de géomorphologie glaciaire

Ecohydrol. Hydrobiol. 4, 379–395. locale dans la chaîne du Bargy (Haute-Savoie). Bulletin de la Société vaudoise

Pinay,G.,Gascuel,C.,Menesguen,A.,Souchon,Y.,LeMoal,M.,Levain,A.,Moatar,F.,Pannard,A., des Sciences naturelles 78, 145–157.

Souchu,P.,2017.L’eutrophisation:manifestations,causes,conséquencesetprédictibilité. Shand, C.A., Wendler, R., 2014. Portable X-ray ﬂuorescence analysis of mineral and

Synthèse de l’Expertise scientiﬁque collective CNRS-Ifremer-INRA-Irstea.. . organic soils and the inﬂuence of organic matter. J. Geochem. Explor.143, 31–42.

Pini, R., Ravazzi, C., Raiteri, L., Guerreschi, A., Castellano, L., Comolli, R., 2017. From doi:http://dx.doi.org/10.1016/j.gexplo.2014.03.005.

pristine forests to high-altitude pastures: an ecological approach to prehistoric Simonneau, A., Doyen, E., Chapron, E., Millet, L., Vannière, B., Di Giovanni, C.,

human impact on vegetation and landscapes in the western Italian Alps. J. Ecol. Bossard, N., Tachikawa, K., Bard, E., Albéric, P., Desmet, M., Roux, G., Lajeunesse,

doi:http://dx.doi.org/10.1111/1365-2745.12767. P., Berger, J.F., Arnaud, F., 2013. Holocene land-use evolution and associated soil

Poulenard, J., Arnaud, F., Perrette, Y., Sabatier, P., Deline, P., Giguet-Covex, C., Bajard, erosion in the French Prealps inferred from Lake Paladru sediments and

M., Mourier, B., Quiers, M., Malet, E., Develle, A.-L., Fanget, B., Faivre, P., archaeological evidences. J. Archaeol. Sci. 40, 1636–1645. doi:http://dx.doi.org/

Dambrine, É., Chalmin, E., 2015. Organic geochemistry of soils and sediments: a 10.1016/j.jas.2012.12.002.

key to describe the anthropocene? Collection EDYTEM - Sols et Matières Sitters, J., Bakker, E.S., Veldhuis, M.P., Veen, G.F., Olde Venterink, H., Vanni, M.J., 2017.

Organiques: Mémoires et Héritages. Presented at the Sessions communes des The stoichiometry of nutrient release by terrestrial herbivores and its

12ème journées d’Etude de Sols - Sols en héritages et de la 2ème réunion des ecosystem consequences. Front. Earth Sci. 5 doi:http://dx.doi.org/10.3389/

chercheurs francophones en géochimie organique, Laboratoire EDYTEM, feart.2017.00032.

Université de Savoie 30 juin - 4 juillet, 2014. , pp. 9–18. Smith, V.H., Tilman, G.D., Nekola, J.C., 1999. Eutrophication: impacts of excess

Punning, J.-M., Kapanen, G., Hang, T., Davydova, N., Kangur, M., 2008. Changes in the nutrient inputs on freshwater, marine, and terrestrial ecosystems. Environ.

water level of Lake Peipsi and their reﬂection in a sediment core. Hydrobiologia Pollut. 100, 179–196.

599, 97–104. doi:http://dx.doi.org/10.1007/s10750-007-9192-0. Steffen, W., Broadgate, W., Deutsch, L., Gaffney, O., Ludwig, C., 2015. The trajectory of

R Development Core Team, 2011. R: A Language and Environment for Statistical the Anthropocene: the great acceleration. Anthr. Rev. 2, 81–98.

Computing. R Foundation for Statistical Computing, Vienna, Austria. Stockmarr, J., 1971. Tablets With Spores Used in Absolute Pollen Analysis. Pollen Et

Reed-Andersen, T., Carpenter, S.R., Lathrop, R.C., 2000. Phosphorus ﬂow in a Spores. .

watershed-lake ecosystem. Ecosystems 3, 561–573. doi:http://dx.doi.org/ Tyson, R., 2012. Sedimentary organic matter: organic facies and palynofacies.

10.1007/s100210000049. Springer Sci. Bus. Media .

Reille, M., 1992. Pollen et spores d’Europe et d’Afrique du Nord. Lab. Bot. Hist. & Valese, E., Conedera, M., Held, A.C., Ascoli, D., 2014. Fire, humans and landscape in

Palynol. URA CNRS/Univ. Marseille III, Marseille. the European Alpine region during the Holocene. Anthropocene 6, 63–74. doi:

Reimer, P.J., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk Ramsey, C., Buck, C. http://dx.doi.org/10.1016/j.ancene.2014.06.006.

E., Cheng, H., Edwards, R.L., Friedrich, M., 2013. IntCal13 and Marine13 Van Geel, B., 2002. Non-pollen palynomorphs. In: Smol, J.P., Birks, H.J.B., Last, W.M.,

Radiocarbon Age Calibration Curves 0-50,000 Years Cal BP. . Bradley, R.S., Alverson, K. (Eds.), Tracking Environmental Change Using Lake

Reyss, J.-L., Schmidt, S., Legeleux, F., Bonté, P.,1995. Large, low background well-type Sediments. Springer, Netherlands, Dordrecht, pp. 99–119. doi:http://dx.doi.org/

detectors for measurements of environmental radioactivity. Nucl. Instrum. 10.1007/0-306-47668-1_6.

Methods Phys. Res. A 357, 391–397. Vitousek, P.M., 1997. Human domination of earth’s ecosystems. Science 277, 494–

Rogora, M., Mosello, R., Arisci, S., 2003. The effect of climate warming on the 499. doi:http://dx.doi.org/10.1126/science.277.5325.494.

hydrochemistry of alpine lakes. Water Air Soil Pollut. 148, 347–361. Wang, Y., Zhang, X., Huang, C., 2009. Spatial variability of soil total nitrogen and soil

Ross, D., Tate, K., Scott, N., Feltham, C., 1999. Land-use change: effects on soil carbon, total phosphorus under different land uses in a small watershed on the Loess

nitrogen and phosphorus pools and ﬂuxes in three adjacent ecosystems. Soil Plateau. China. Geoderma 150, 141–149. doi:http://dx.doi.org/10.1016/j.

Biol. Biochem. 31, 803–813. geoderma.2009.01.021.

Sabatier, P., Dezileau, L., Briqueu, L., Colin, C., Siani, G., 2010. Clay minerals and Wetzel, R.G., 2001. Limnology: Lake and River Ecosystems. Gulf Professional

geochemistry record from northwest Mediterranean coastal lagoon sequence: Publishing.

Implications for paleostorm reconstruction. Sediment. Geol. 228, 205–217. doi: Williams, P.H., Haynes, R.J., 1990. Inﬂuence of improved pastures and grazing

http://dx.doi.org/10.1016/j.sedgeo.2010.04.012. animals on nutrient cycling within New Zealand soils. N. Z. J. Ecol. 49–57.

Sabatier, P., Poulenard, J., Fanget, B., Reyss, J.-L., Develle, A.-L., Wilhelm, B., Ployon, E., Williamson, C.E., Dodds, W., Kratz, T.K., Palmer, M.A., 2008. Lakes and streams as

Pignol, C., Naffrechoux, E., Dorioz, J.-M., others, 2014. Long-term relationships sentinels of environmental change in terrestrial and atmospheric processes.

among pesticide applications, mobility, and soil erosion in a vineyard Front. Ecol. Environ. 6, 247–254.

watershed. Proc. Natl. Acad. Sci. 111, 15647–15652. WRB - FAO, 2014. World Reference Base for Soil Resources 2014 International Soil

Saether, O.A.,1979. Chironomid communities as water quality indicators. Ecography Classiﬁcation System for Naming Soils and Creating Legends for Soil Maps. FAO,

2, 65–74. Rome.